Yasheng Chen1, Chunwei Ying2, Michael M Binkley1, Meher R Juttukonda3,4, Shaney Flores5, Richard Laforest5, Tammie L S Benzinger5, Hongyu An1,2,5. 1. Department of Neurology, Washington University School of Medicine in St. Louis, St. Louis, Missouri, USA. 2. Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, Missouri, USA. 3. Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, Massachusetts, USA. 4. Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA. 5. Mallinckrodt Institute of Radiology, Washington University School of Medicine in St. Louis, St. Louis, Missouri, USA.
Abstract
PURPOSE: The accuracy of existing PET/MR attenuation correction (AC) has been limited by a lack of correlation between MR signal and tissue electron density. Based on our finding that longitudinal relaxation rate, or R1 , is associated with CT Hounsfield unit in bone and soft tissues in the brain, we propose a deep learning T1 -enhanced selection of linear attenuation coefficients (DL-TESLA) method to incorporate quantitative R1 for PET/MR AC and evaluate its accuracy and longitudinal test-retest repeatability in brain PET/MR imaging. METHODS: DL-TESLA uses a 3D residual UNet (ResUNet) for pseudo-CT (pCT) estimation. With a total of 174 participants, we compared PET AC accuracy of DL-TESLA to 3 other methods adopting similar 3D ResUNet structures but using UTE R 2 ∗ , or Dixon, or T1 -MPRAGE as input. With images from 23 additional participants repeatedly scanned, the test-retest differences and within-subject coefficient of variation of standardized uptake value ratios (SUVR) were compared between PET images reconstructed using either DL-TESLA or CT for AC. RESULTS: DL-TESLA had (1) significantly lower mean absolute error in pCT, (2) the highest Dice coefficients in both bone and air, (3) significantly lower PET relative absolute error in whole brain and various brain regions, (4) the highest percentage of voxels with a PET relative error within both ±3% and ±5%, (5) similar to CT test-retest differences in SUVRs from the cerebrum and mean cortical (MC) region, and (6) similar to CT within-subject coefficient of variation in cerebrum and MC. CONCLUSION: DL-TESLA demonstrates excellent PET/MR AC accuracy and test-retest repeatability.
PURPOSE: The accuracy of existing PET/MR attenuation correction (AC) has been limited by a lack of correlation between MR signal and tissue electron density. Based on our finding that longitudinal relaxation rate, or R1 , is associated with CT Hounsfield unit in bone and soft tissues in the brain, we propose a deep learning T1 -enhanced selection of linear attenuation coefficients (DL-TESLA) method to incorporate quantitative R1 for PET/MR AC and evaluate its accuracy and longitudinal test-retest repeatability in brain PET/MR imaging. METHODS: DL-TESLA uses a 3D residual UNet (ResUNet) for pseudo-CT (pCT) estimation. With a total of 174 participants, we compared PET AC accuracy of DL-TESLA to 3 other methods adopting similar 3D ResUNet structures but using UTE R 2 ∗ , or Dixon, or T1 -MPRAGE as input. With images from 23 additional participants repeatedly scanned, the test-retest differences and within-subject coefficient of variation of standardized uptake value ratios (SUVR) were compared between PET images reconstructed using either DL-TESLA or CT for AC. RESULTS: DL-TESLA had (1) significantly lower mean absolute error in pCT, (2) the highest Dice coefficients in both bone and air, (3) significantly lower PET relative absolute error in whole brain and various brain regions, (4) the highest percentage of voxels with a PET relative error within both ±3% and ±5%, (5) similar to CT test-retest differences in SUVRs from the cerebrum and mean cortical (MC) region, and (6) similar to CT within-subject coefficient of variation in cerebrum and MC. CONCLUSION: DL-TESLA demonstrates excellent PET/MR AC accuracy and test-retest repeatability.
Authors: V Schulz; I Torres-Espallardo; S Renisch; Z Hu; N Ojha; P Börnert; M Perkuhn; T Niendorf; W M Schäfer; H Brockmann; T Krohn; A Buhl; R W Günther; F M Mottaghy; G A Krombach Journal: Eur J Nucl Med Mol Imaging Date: 2010-10-05 Impact factor: 9.236
Authors: Vincent Keereman; Yves Fierens; Tom Broux; Yves De Deene; Max Lonneux; Stefaan Vandenberghe Journal: J Nucl Med Date: 2010-05 Impact factor: 10.057
Authors: E K Fram; R J Herfkens; G A Johnson; G H Glover; J P Karis; A Shimakawa; T G Perkins; N J Pelc Journal: Magn Reson Imaging Date: 1987 Impact factor: 2.546
Authors: Ciprian Catana; Andre van der Kouwe; Thomas Benner; Christian J Michel; Michael Hamm; Matthias Fenchel; Bruce Fischl; Bruce Rosen; Matthias Schmand; A Gregory Sorensen Journal: J Nucl Med Date: 2010-09 Impact factor: 10.057
Authors: Tri Huynh; Yaozong Gao; Jiayin Kang; Li Wang; Pei Zhang; Jun Lian; Dinggang Shen Journal: IEEE Trans Med Imaging Date: 2015-07-28 Impact factor: 10.048
Authors: Paul Blanc-Durand; Maya Khalife; Brian Sgard; Sandeep Kaushik; Marine Soret; Amal Tiss; Georges El Fakhri; Marie-Odile Habert; Florian Wiesinger; Aurélie Kas Journal: PLoS One Date: 2019-10-07 Impact factor: 3.240
Authors: Parna Eshraghi Boroojeni; Yasheng Chen; Paul K Commean; Cihat Eldeniz; Gary B Skolnick; Corinne Merrill; Kamlesh B Patel; Hongyu An Journal: Magn Reson Med Date: 2022-06-17 Impact factor: 3.737
Authors: Eric Guedj; Andrea Varrone; Ronald Boellaard; Nathalie L Albert; Henryk Barthel; Bart van Berckel; Matthias Brendel; Diego Cecchin; Ozgul Ekmekcioglu; Valentina Garibotto; Adriaan A Lammertsma; Ian Law; Iván Peñuelas; Franck Semah; Tatjana Traub-Weidinger; Elsmarieke van de Giessen; Donatienne Van Weehaeghe; Silvia Morbelli Journal: Eur J Nucl Med Mol Imaging Date: 2021-12-09 Impact factor: 10.057